Optimizing IoT Network Resource Management Through Machine Learning-Based Bandwidth Prediction.
DOI:
https://doi.org/10.64479/iarci.v2i1.82Keywords:
Internet of Things (IoT), Bandwidth Allocation Prediction, Machine Learning, Network Traffic, Network Resource ManagementAbstract
The rapid growth of Internet of Things (IoT) technologies has increased the complexity of network resource management, particularly bandwidth allocation. Efficient allocation is essential for maintaining communication reliability, minimizing latency, and ensuring quality of service (QoS) in heterogeneous IoT environments. This study analyzes the relationship between network traffic characteristics and allocated bandwidth using a machine learning approach. The publicly available IoT Network Traffic Dataset contains 100,000 communication records, including packet size, transmission time, protocol type, bandwidth usage, latency, jitter, packet loss rate, device type, and energy usage. Data preprocessing involved cleaning, categorical encoding, feature scaling, and exploratory data analysis. Random Forest Regression was used to predict allocated bandwidth from these characteristics. Feature importance analysis identified transmission time, energy usage, bandwidth usage, packet loss rate, latency, and jitter as the most influential predictors. However, the model achieved training and testing R² scores of 0.045 and −0.003, respectively, indicating limited predictive capability and little predictive signal captured by the model. These findings highlight the importance of feature relevance and dataset quality in developing intelligent network resource management systems. Future research should incorporate additional network-related features and evaluate alternative machine learning models to improve bandwidth allocation prediction in large-scale IoT environments.
Downloads
References
[1] S. M. Basha, R. Dhanalakshmi, A. Soundari, V. C. B, and B. Rajasekar, “An Efficient Allocation of Resources in wireless Communications IoT Network for Numerous Users.” 2023. doi: 10.1109/ICSES60034.2023.10465489.
[2] I. Chakour, C. Daoui, M. Baslam, B. Sainz-De-Abajo, and B. Garcia-Zapirain, “Strategic Bandwidth Allocation for QoS in IoT Gateway: Predicting Future Needs Based on IoT Device Habits.” 2024. doi: 10.1109/ACCESS.2024.3351111.
[3] K. Saurabh, S. Mahapatra, and M. M. Tripathi, “An Adaptive Bandwidth Scheduling Algorithm for IoT Devices Using Machine Learning Techniques.” 2026. doi: 10.1016/j.fraope.2026.100492.
[4] A. Hameed, J. Violos, A. Leivadeas, N. Santi, R. Grünblatt, and N. Mitton, “Toward QoS Prediction Based on Temporal Transformers for IoT Applications.” 2022. doi: 10.1109/TNSM.2022.3217170.
[5] J. D. Roselind, H. Naveen, and S.Yuvaraj, “Artificial Intelligence enabled Dynamic Bandwidth Allocation Optimization in IoT for improved QoS.” 2023. doi: 10.58599/ijsmien.2023.1204.
[6] A. Alzaidy, “Real-Time Bandwidth Allocation Strategy for Next-Generation Heterogeneous Network Infrastructures.” 2025. doi: 10.5220/0014193200004932.
[7] I. Chakour, C. Daoui, M. Baslam, B. Sainz-De-Abajo, and B. Garcia-Zapirain, “Strategic Bandwidth Allocation for QoS in IoT Gateway: Predicting Future Needs Based on IoT Device Habits.” 2024. doi: 10.1109/ACCESS.2024.3351111.
[8] Shekhar, G. Bhandari, and S. Tyagi, “Enhance Quality of Services in Internet of Things using Improved Dynamic Bandwidth Allocation (IDBA) Techniques.” 2025. doi: 10.1016/j.procs.2025.04.082.
[9] B. Lee, “Efficient Resource Management for Massive MIMO in High-Density Massive IoT Networks.” 2025. doi: 10.1109/TMC.2024.3486712.
[10] M. N and V. K, “Enhanced Dynamic Bandwidth Allocation (EDBA) Method for Improved Quality of Service in IoT Systems.” 2026. doi: 10.1109/ICMCSI67283.2026.11412892.
[11] M. Manu et al., “Optimizing Quality of Service in IoT PORTAL through Strategic Bandwidth Allocation: A NEXUS access model.” 2024. doi: 10.1109/ICACITE60783.2024.10617219.
[12] S. Tidjani, Z. Hammoudi, and M. E. Moad, “Low complexity multichannel spectrum prediction algorithm based on optimized neural network for spectrum allocation in cognitive radio internet of things.” 2022. doi: 10.1002/ett.4562.
[13] N. Wang, Y. Liu, Y. Zhou, Z. Li, Z. Li, and Q. Song, “Distribution Internet of Things Local Communication Bandwidth Prediction.” 2022. doi: 10.1109/ICSMD57530.2022.10058402.
[14] A. Chen, J. Liu, and J. Li, “Multi-Mode Data Prediction-Based Dynamic Bandwidth Allocation for IoT-Empowered Electric Equipment Monitoring.” 2025. doi: 10.1049/cmu2.70098.
[15] W. Symbor and Ł. Falas, “Ensuring Reliable Network Communication and Data Processing in Internet of Things Systems with Prediction-Based Resource Allocation.” 2025. doi: 10.3390/s25010247.
[16] A. Hameed, J. Violos, N. Santi, A. Leivadeas, and N. Mitton, “FeD-TST: Federated Temporal Sparse Transformers for QoS Prediction in Dynamic IoT Networks.” 2025. doi: 10.1109/TNSM.2024.3493758.
[17] C. Awanyo and N. Guermouche, “Deep Neural Network-Based Approach for IoT Service QoS Prediction.” 2023. doi: 10.1007/978-981-99-7254-8_31.
[18] S. Tidjani, “An Efficient Predictive Spectrum Allocation Algorithm for Multi-User Cognitive Radio IoT Devices Within GSM900 Band.” 2024. doi: 10.1109/ICEET65156.2024.10913610.
[19] Nathaniel Bastian; David Bierbrauer; Morgan McKenzie; Emily Nack, “Aci iot network traffic dataset 2023.” IEEE Dataport, 2023. [Online]. Available: 10.21227/qacj-3x32
[20] V. Sharma, R. Arya, and S. Kumar, “Efficient Channel Prediction Technique Using AMC and Deep Learning Algorithm for 5G (NR) mMTC Devices.” 2022. doi: 10.1109/ACCESS.2022.3167442.
[21] K. W. Cho, M. Cominelli, F. Gringoli, J. Widmer, and K. Jamieson, “Scalable Multi-Modal Learning for Cross-Link Channel Prediction in Massive IoT Networks.” 2023. doi: 10.1109/TON.2025.3553023.
